All prominent ethereum second layer (L2) projects have gathered in one stage for the first time in Amsterdam at an ongoing two-day technical conference dedicated to Ethereum L2 Scaling.
Representatives from Arbitrum, ZKSync, Optimism, StarkWare and Fuel (formerly Celestia), all pictured in that order, spoke of the challenges faced as well as potential timelines.
All stated that while they are decentralized in as far as censorship resistance and funds are secured by the ethereum base layer, there are tradeoffs in regards to the upgradability of the contract.
They also use similar governance methods in having some sort of input by different stakeholders while balancing quick upgrades in cases of bugs with allowing users to exit the network either because they don’t agree with the upgrade or for whatever reason.
Somewhat newcomer Fuel however stated they don’t upgrade at all, instead they follow the Uniswap approach in V1, V2, but Arbitrum’s co-founder Ed Felten pointed out that you need to upgrade to keep up with ethereum forks.
John Adler, a former ConsenSys alumni and co-founder of Fuel Labs, also suggested that once data sharding is implemented in eth, there can be what he calls Sovereign Rollups.
That is a second layer that uses ethereum only for data availability, with that second layer then able to soft and hard fork.
In regards to a timeline, none gave a date, focusing on milestones. For Arbitrum, the focus is on opening up validation which should be happening “quite soon.”
Then they need to decentralize the sequencer which Felton said can’t censor, but can manipulate the order of transactions. That will become a distributed sequencer in Arbitrum, with this being the last change they make out of all milestones.
For zkSync, they need to get out zksync v2 which has smart contracts, and then zkPorter, a seperate data availability layer.
zkPorter will need a token, with the focus then being on decentralizing zkPorter validators.
For Alex Gluchowski of zkSync, the sequencer can censor specific transactions and even dapps, so they need to decentralize that by introducing a layer two consensus.
Finally they need to decentralize the proof generation. As this works on zero knowledge cryptography, transactions are condensed into a hash with a proof – generated by anyone – allowing you to verify them. That proof generation can be computing intensive, so they plan to decentralize it by allowing anyone with a GPU to generate it.
Finally, Ben Jones of Optimism has as a milestone the decentralization of the development of the protocol. He spoke of there being different implementations of Optimism, like there are different clients for ethereum. If there are bugs, then you can upgrade the specific implementation, rather than the whole network, so increasing overall resiliance.
This fairly short session is more notable for them all being together in a sign that a new branch of ethereum is forming.
Second layers, though in development since 2018, used to be more just one presentation at a conference about future projects.
Now they have a whole conference of their own, with one of them potentially to be the future of ethereum scaling, if not maybe even all of them.
Because the base layer won’t scale anytime soon. There will be some additions in regards to sharding data, but much is suggested will run in one of these projects.
They hold a combined $6.5 billion worth of assets currently, or some two million eth. That’s circa 2% of all eth in circulation.
So there’s a long way to go for one of these projects to be the answer to scaling, with two of them – zkSync and Starknet – not even having a mainnet end-users EVM yet, while Arbitrum and Optimism still have training wheels.
Yet just in September they had 200,000 eth, making it a 10x in months at such an early stage which suggests there’s significant demand for layered scaling and that development is moving.
